Dielectric Function Data for Wide-Gap Materials Computed with PBE and TASK

3.8 GB of dielectric function data generated for a materials screening study. The data includes traces of dielectric functions computed with PBE and TASK methods, sampled at 0.1 eV intervals from 0 to 20 eV. The dataset was created by Pedro Borlido and colleagues, with funding from FCT, and was last updated in May 2026.

Use Cases

• Training machine learning models for dielectric function prediction based on material composition.
• Screening for high refractive index materials using the provided frequency-dependent dielectric data.
• Benchmarking computational methods (PBE vs. TASK) for calculating optical properties.
• Expanding libraries of dielectric functions by integrating data from other sources, as suggested by the authors.

Strengths

• Data is sampled at a high resolution of 0.1 eV intervals across a 0-20 eV range.
• Includes both original and smoothed versions of the PBE-calculated dielectric functions.
• Comes with associated scripts and trained model parameters (ALIGNN+AE) for reproducibility.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• The authors note that silent errors may have passed through their checks.

Provenance

Source

Pedro Borlido, CFisUC

Collection Method

Generated via computational physics simulations using Quantum Espresso and other codes.

Freshness

Last updated 2026-05-27 15:04:56